209 replies to this topic

[Makaru]

You and I either think the same... or you stole that Idea from me. Either way I'm glad it's getting attention because it needs it.

[Big Tin Man]

wanderer, on 15 September 2015 - 02:01 PM, said:

Given, this means long-range weapons actually might end up "splitting" before hitting the end of their effective range, which could lead to misses. I'd rather not see that happening with "default" firing and would set it further out to "maximum range weapons deal damage". That's what, 2500m or so?

edit: TL;DR--unlocked target convergence of any kind serves to benefit 'alphatardwarriors' shooting at long range. Followed by numbers.

---------------------------------------------------------------------------------------

I think you're missing the issue with having any sort of convergence for an unlocked target. We're talking about mechs that are about 5 meters wide, so it's not that hard to make a slight offset adjustment and set your weapon groups in a left/right config and adjust your aim.

I'm still against having any default convergence point, as it gives the tryhards a magic range where they still fairly effective. For example, let's say weapons converge at 2000m, and I will normalize mechs into a percentage of a mech's width from left arm to right arm (yes, I'm assuming mech's weapons are equally distributed across its width).

Weapon spread///distance
100% of mech width /// 0 meters
50% of mech width /// 1000 meters
0% of mech width (pinpoint) /// 2000 meters

Now let's min/max it and go for an asymmetric load out, assuming only CT, LT and LA are used (or RT, CT, LT)

50% of mech width /// 0 meters
25% of mech width /// 1000 meters
0% of mech width (pinpoint) /// 2000 meters

Even more asymmetric, only LT and LA

25% of mech width /// 0 meters
12.5% of mech width /// 1000 meters
0% of mech width (pinpoint) /// 2000 meters

There is a sweet spot around 1,000 meters for an asymmetric or torso only loadout, where they're still doing full damage, and hitting with a tight enough group that could still focus a particular section of a mech. Convergence should be set at infinity, or else a meta will be built around exploiting the unlocked target convergence.

Edited by Big Tin Man, 15 September 2015 - 04:27 PM.

[GreyNovember]

Okay, let's assume this is seriously being looked at as a mechanic.

What would your crosshairs look like, then?

If I had a nova with 12 lasers, am I going to see 12 dots around my main crosshair, and when I lock on, will they all start converging?

[wanderer]

GreyNovember, on 15 September 2015 - 05:03 PM, said:

Okay, let's assume this is seriously being looked at as a mechanic.

What would your crosshairs look like, then?

If I had a nova with 12 lasers, am I going to see 12 dots around my main crosshair, and when I lock on, will they all start converging?

They'll look the same way as before. Why would they need to change? You either have a red box to lock onto (and if so, it's instantly converging at that range) or you don't (and your guns are firing at the default long-distance convergence).

[GreyNovember]

wanderer, on 15 September 2015 - 05:43 PM, said:

They'll look the same way as before. Why would they need to change? You either have a red box to lock onto (and if so, it's instantly converging at that range) or you don't (and your guns are firing at the default long-distance convergence).

And as I understand it, this default long distance convergence is dependent on the weapon being fired, yes?

So if I had say 6ERML and 6ERSL, those would converge at their maximum ranges when unlocked, and they will function as they do now if I do have a lock; regardless of ECM interference; so long as I have a lock?

Will it take time to converge? If it does, what indicator do we have that the weapons are aligned/not aligned? Hence the question about multiple crosshairs.

[Wibbledtodeath]

WOW. What a great Idea!!!!

Not that most ppl will be reading a post 8 pages in but its worth a post if nothing else than to bump a great concept.

Only problem I can see is if converging on a "point" at max sensor range, or weapon range the weapon fire will make an elongated X.

Rather, weapons should not converge at all (just fire forward in parallel/infinity convergence) unless there is a sensor lock (and that is at all ranges- so a long range sniper mech may have trouble converging on a fast nearby target also without additional scouting help). This would make tight weapon groupings on a chassis a more important balancing factor re hard point location and actually effectively reduce the oh god fire power of many bigger boating PPFLD mechs- to an extent/when not locked/scouted (I will champion just about anything that helps balance and eliminates need for ghost heat).

Using "lock" to converge will encourage people to fire at the one target- good practice anyway- and it would encourage/reward team play. But may be frustrating in PUG matches (if you are the unlucky one spotted/targeted by a scout, or if your Dire whale gets abandoned by your team). Possibly tying convergence to all mechs ELIGIBLE to be targeted would work (still requiring scouting, but unlike LRMs not requiring lock). You could also make this "eligibility" criteria ignore ECM (but still protecting the ECM'd mechs from full target locks) to prevent ECM being the god shield vs everything.

Alternatively- tie "eligibility" targeting for to equipment (e.g. command chair, TC), have everyone else only scout the mech they have a lock on?

Edited by Wibbledtodeath, 15 September 2015 - 06:14 PM.

[Livewyr]

Long Live ECM!

...

[wanderer]

GreyNovember, on 15 September 2015 - 05:59 PM, said:

And as I understand it, this default long distance convergence is dependent on the weapon being fired, yes?

So if I had say 6ERML and 6ERSL, those would converge at their maximum ranges when unlocked, and they will function as they do now if I do have a lock; regardless of ECM interference; so long as I have a lock?

Will it take time to converge? If it does, what indicator do we have that the weapons are aligned/not aligned? Hence the question about multiple crosshairs.

The choice is actually multiple: The simplest default would be the maximum distance possible, period. I tend to prefer "maximum distance -any- weapon can inflict damage" (the exact number after modules and the like escapes me, but I believe it's around 2000m). Others have suggested shorter distances. The most complex would be "all weapons converge by default at the maximum distance they inflict damage"- which given the fits that may put the MWO engine through is also the least likely to be implementable.

ECM will change nothing if you have a lock (though IMHO, ECM should change to "target is treated as being further away for the purposes of sensors" rather than "Poof, I'm invisible ninja!"). At that point, all weapons go from "default" to "converge at the range the locked-on target is at" immediately, regardless of if they can actually reach that distance and firing at the point that is that distance away from whatever direction you've got the crosshairs. Of course, if you're aiming at the locked target, obviously that means you're hitting it with near-perfect convergence- anything else will have to deal with the resulting too-close/too-long effects of not being at the right distance when struck, and of course if you don't have your crosshairs on the locked target, you're not hitting it. Lock-on gives your guns the range to target, but it doesn't make your lasers turn right angles. It just gives them convergence at the right range to focus all your shots on the same point. Missiles operate as they do now.

Wibbledtodeath, on 15 September 2015 - 06:06 PM, said:

WOW. What a great Idea!!!!

Not that most ppl will be reading a post 8 pages in but its worth a post if nothing else than to bump a great concept.

Only problem I can see is if converging on a "point" at max sensor range, or weapon range the weapon fire will make an elongated X.

What's the problem? That's what happens to a converging salvo of fire when it travels past the distance it's set to converge at. That is, targets closer than the point of or further from it will suffer increasingly spread damage, and I fully expect people missing locked targets and having their barrage first gather from wide spread to single point and back outwards again.

Quote

Rather, weapons should not converge at all (just fire forward in parallel/infinity convergence) unless there is a sensor lock (and that is at all ranges- so a long range sniper mech may have trouble converging on a fast nearby target also without additional scouting help). This would make tight weapon groupings on a chassis a more important balancing factor re hard point location and actually effectively reduce the oh god fire power of many bigger boating PPFLD mechs- to an extent/when not locked/scouted (I will champion just about anything that helps balance and eliminates need for ghost heat).

I'd like to think that a 31st century combat vehicle at least has some reasonable default convergence distance, considering such things have existed since WWII. That is, a pilot can expect SOME minimum level of clustering hits on their target. Imperfect converging shots as a result of this should scatter damage enough to matter without causing whiff-fests, certainly in most cases enough to keep anything not on the same location from getting a good same-location hit at range.

And most of the truly most heinous PPFLD boats don't put their Death Star on a single location.

Quote

Using "lock" to converge will encourage people to fire at the one target- good practice anyway- and it would encourage/reward team play. But may be frustrating in PUG matches (if you are the unlucky one spotted/targeted by a scout, or if your Dire whale gets abandoned by your team). Possibly tying convergence to all mechs ELIGIBLE to be targeted would work (still requiring scouting, but unlike LRMs not requiring lock). You could also make this "eligibility" criteria ignore ECM (but still protecting the ECM'd mechs from full target locks) to prevent ECM being the god shield vs everything.

Alternatively- tie "eligibility" targeting for to equipment (e.g. command chair, TC), have everyone else only scout the mech they have a lock on?

Nope. You want convergence, you actually target a 'Mech, you get the red box, you shoot. Anything else gets you default convergence. Not "could be targeted" (seriously, that's everything in sensor range of any friendly, which may as well be the status quo). Not "ignoring ECM" (because Jesusbox needs to be fixed or info warfare is a moot point. Praise Helix.).

Command consoles and TC's can figure into this, but to me that's beyond the basic mechanics of the OP.

Edited by wanderer, 15 September 2015 - 09:53 PM.

[Tahawus]

To try to sum up: (because I'm waiting for statistics to run)

1. Reinstate convergence to MWO
2. Default convergence at infinity (i.e. parallel straight ahead).
3. Weapons only converge when the victim is targeted.

Reasons this is good.
1. It makes information/targeting by the team critical to success. (you wanted information warfare, you got it).
2. It cuts down on long range high precision sniping and poptarting/hillhumping (unless there's a spotter, then go for it).
3. Opens the possibility for skill tree (re)development on the convergence and info warfare tracks.
4. (my opinon) will likely cut down on P2W(v.2), where people with high spec machines can see you before you see them and accurately hit you at near max range.

Things it depends on:
1. Netcode
2. ECM revisions

Related:
I think this approach would really benefit from the information warfare quirks that PGI just mishandled so badly in the PTS.
Similarly, some mechs could have convergence quirks, and others penalties. Think Jaeger with limited arm actuators and ac20s with a light mech circulating that can't converge to bring both to bear simultaneously), not so scary now are we....

I actually think it would be really nice to have one reticule for each portion of the body that gives a visual clue on convergence, and would let you turn loose with the weapons on one arm or the other without full convergence as a middle ground.

[Hans Von Lohman]

GreyNovember, on 15 September 2015 - 05:59 PM, said:

And as I understand it, this default long distance convergence is dependent on the weapon being fired, yes?

So if I had say 6ERML and 6ERSL, those would converge at their maximum ranges when unlocked, and they will function as they do now if I do have a lock; regardless of ECM interference; so long as I have a lock?

Will it take time to converge? If it does, what indicator do we have that the weapons are aligned/not aligned? Hence the question about multiple crosshairs.

No, the range would probably be set to infinity. In other words all of your weapons would make a shot pattern in the shape of where the hardpoints are on your mech. I.E. a Dual Guass Rifle Jaeger with nothing else will have a gauss rifle shot that hits to the left of your gunsight, and another to the right, one from each arm. Neither will hit the center, ever, at any range until you target something.

There is no time to converge your weapons right now. There used to be a delay in convergence in beta, but that got removed. We still have the pointless Elite pilot skill called PinPoint to speed up the process, but nobody cares now because that actually happens instantly, aka 0.00 seconds to reset your convergence. You can't make it go any faster than that, despite the fact that your Elite Pilot skill makes it try to go faster.

What we are saying is this.

Imagine that your convergence point where all of your weapons focus onto will either be the range of your target, or infinity aka no convergence at all if you don't have a target.

Lets assume you own a Hunchback-4G with a stock loadout. Without a lock onto anything your AC-20 round will hit just a bit to the right of your crosshair (the same distance from your cockpit to the gun barrel), your head small-laser will just a tiny bit above the crosshair, and each arm medium-laser will hit below and to the side of the crosshair.

At all ranges, that same shot pattern will happen.

However, if you get a lock onto an enemy, then instantly, with no delay, all of your lasers and gun will converge on the range that target is to you, just like it already does. If that enemy Atlas is just standing there, AFK or disconnected, or in the training maps, you can zoom in on his left eye cockpit window, alpha strike, and everything will hit, probably killing the Atlas instantly (boom, head-shot).

If you didn't lock onto the Atlas before alpha striking him, your head laser will probably be the only thing that hits the left eye/cockpit. Your medium lasers will probably hit the Atlas in his CT and Left Arm, and your AC-20 will wiz past the left ear of the Atlas head and into thin air.

Edited by Hans Von Lohman, 15 September 2015 - 10:26 PM.

[Big Tin Man]

From pages back:

GreyNovember, on 15 September 2015 - 05:03 PM, said:

What would your crosshairs look like, then?

If I had a nova with 12 lasers, am I going to see 12 dots around my main crosshair, and when I lock on, will they all start converging?

Big Tin Man, on 14 September 2015 - 03:35 PM, said:

Hmmm, you got me thinking about a dynamic reticule (not sure if this is possible, I recall it being discussed before). Something like this:

Unlocked, no convergence mode:

o >-------< o

[where the o's are the left and right arm, and the X part of the crosshair is widened to represent the mech's torso width]

And it would morph into the current crosshair when targeting something

>o<

And Wanderer, how about a compromise on the convergence question. Unleveled mechs will not converge (converge at infinity) on unlocked targets, the pinpoint skill will set a convergence point at max damage range for unlocked targets? I still think this will further push an asymmetric build meta, but it will push a lot of change for the better.

[Prof RJ Gumby]

I would love to see that idea tested on PTS with current quirks and/or PTS targeting quirks.

It sounds promising, but there it brings many implications (some already voiced in the newer posts).

1. Buff to small mechs, nerf to big mechs.
Small mechs have their hardpoints so close they would often be able to hit bigger mechs in one component without lock, while it would be impossible for the bigger mechs to return that favour. 'Unlocked' lights would be near invincible, while unlocked assaults, especially those big ones, could be wrecked by any smaller mechs with ease. Large width would be even more of a disadvantage than it is now. Think of the king crab and how big his enemy would need to be to actually care for locking and convergence. Most wouldn't.

2. If convergence is set on a certain distance, there will be builds to exploit that
Max sensor range convergence? Can you imagine all those ERLL snipers owning everybody else? You can do proper damage at 'untargeted convergence point', you own. Thus there should be no point of convergence at all to have a chance of this idea not be exploited as heck, just 'moderately exploited' (more in point 3)

3. Grouped hardpoints > distanced hardpoints.
Somebody here voiced his concerns about that wide hardpoints on Nova. It's totally unfounded IMHO, the way I see it, the Nova would be awesome, if not blatantly OP. At least against big targets. All you would need to do is learn to compensate for the spread - aim left arm weapons a bit to the right and right arm weapons a bit to the left. Voila! You have up to 6 lasers hitting almost the exact spot regardless of locks!
Same with some snipers, like the ERLL raven. 2-3 erll lasers grouped together nicely, aim a bit to the left depending on the distance and you have nearly perfect convergence at any ranges in any circumstances. CT weapons would also get a big buff, as they don't need to coverge at all.

Now imagine the poor Blackjack or Jager. Not the laser vomit BJ one, this one would do ok I think. Imagine any other blackjack or Jager. Bah!, imagine any other mech that have hardpoints spread all over the place. Then think what quirks such mechs would need to compensate for that total lock-dependence in comparison to the few mechs who (because of location of hardpoints) are less punished (or go completely unpunished) by the 'no-lock-no-convergence' system.

EDIT:

4. Buff to lasers
Lasers are the only weapons that show you where you shoot exactly - that beam would be very informative. It would allow you to realign your aim while you shoot so that most lasers will hit the component you want to damage. No other weapons would allow that, at least to such extent.
Imagine Nova again. Shoot one laser from right arm, correct aim so it goes when you want it to go, then shoot all the other lasers from that arm. Voila.

*Now I don't own a Nova, so there is a slim chance I overestimate the closeness of its arm hardpoints (they are in circle around the arm as far as I remember). However, even if there is a mistake on my part when it comes to Nova, there are enough closely located hardpoints on many mechs that could be used to exploit the system.

Edited by Prof RJ Gumby, 16 September 2015 - 12:40 AM.

[pyrocomp]

Tahawus, on 15 September 2015 - 09:53 PM, said:

To try to sum up: (because I'm waiting for statistics to run)

1. Reinstate convergence to MWO
2. Default convergence at infinity (i.e. parallel straight ahead).
3. Weapons only converge when the victim is targeted.

Reasons this is good.
1. It makes information/targeting by the team critical to success. (you wanted information warfare, you got it).
2. It cuts down on long range high precision sniping and poptarting/hillhumping (unless there's a spotter, then go for it).
3. Opens the possibility for skill tree (re)development on the convergence and info warfare tracks.
4. (my opinon) will likely cut down on P2W(v.2), where people with high spec machines can see you before you see them and accurately hit you at near max range.

Things it depends on:
1. Netcode
2. ECM revisions

Related:
I think this approach would really benefit from the information warfare quirks that PGI just mishandled so badly in the PTS.
Similarly, some mechs could have convergence quirks, and others penalties. Think Jaeger with limited arm actuators and ac20s with a light mech circulating that can't converge to bring both to bear simultaneously), not so scary now are we....

I actually think it would be really nice to have one reticule for each portion of the body that gives a visual clue on convergence, and would let you turn loose with the weapons on one arm or the other without full convergence as a middle ground.

I agree on reinstating convergence, but weapons should by default converge not in infinity, but at optimal distance. More logical and somewhat affects boating on a single component assorted weapons. GH is still with us, so seeing cERLL plus cERMLs on a single arm is not uncommon. Convergence to different points will affect that even if weapons are clustered on a mech.

Oh, and netcode is not a problem, everything is already in place. ECM however will need some rework. E.g. it should only affect the mech in it's radius. Low radar profile from long range is another thing like nullsig armor (possibly a place to revive FF being less radar bright).

[Duke Nedo]

I would much prefer to have convergence at some distance as fallback value, if so I could support this idea. I am not 100% convinced, but it's not a bad idea. All I know though is that it would simply be annoying like hell to be forced to fire parllell shots like a shotgun at a target just because you can't lock him. Especially annoying would be to deal with ECM-RVN-3L ERLL snipers for example.

Also, piloting mechs with lots of torso hardpoints and AP would be mandatory. Poor mechs with all hardpoints in the arms... poor mechs with very spread out hardpoints...

It may be worth a try though.

[Greyhart]

WOW an idea that the vast majority of people like. It's simple to understand. It makes targeting important. Makes sniping more difficult. Gives lights and TAG and NARC a boost.

Just the 8 pages of people agreeing should mean (if there is any sense) that this should be tested (can it be that hard to code?).

I fear though that it is a further sign of the coming apocalypse, the end times are upon us.

[Livewyr]

Greyhart, on 16 September 2015 - 06:01 AM, said:

WOW an idea that the vast majority of people like. It's simple to understand. It makes targeting important. Makes sniping more difficult. Gives lights and TAG and NARC a boost.

Just the 8 pages of people agreeing should mean (if there is any sense) that this should be tested (can it be that hard to code?).

I fear though that it is a further sign of the coming apocalypse, the end times are upon us.

If you don't hate MWO ECM now... you would if this got implemented.

[rook]

wanderer, on 15 September 2015 - 09:39 PM, said:

The choice is actually multiple: The simplest default would be the maximum distance possible, period. I tend to prefer "maximum distance -any- weapon can inflict damage" (the exact number after modules and the like escapes me, but I believe it's around 2000m). Others have suggested shorter distances. The most complex would be "all weapons converge by default at the maximum distance they inflict damage"- which given the fits that may put the MWO engine through is also the least likely to be implementable.

Couldn't you also put the convergence distance super close, something like 100m? Weapons would still diverge at far distances.

[DivineEvil]

Greyhart, on 16 September 2015 - 06:01 AM, said:

WOW an idea that the vast majority of people like. It's simple to understand. It makes targeting important. Makes sniping more difficult. Gives lights and TAG and NARC a boost.

Just the 8 pages of people agreeing should mean (if there is any sense) that this should be tested (can it be that hard to code?).

I fear though that it is a further sign of the coming apocalypse, the end times are upon us.

Don't rush your assumtions. It might be that just as me, there's people who doesn't feel right to involve themselves into that mess of unreasonably complex ideas.

I'm all for convergence mechanics - just another way to balance mechs around each-other. Nevertheless, my weapons must converge at whatever I'm pointing at no matter what it is. Maybe it might take a couple of seconds, maybe my fire will not be as accurate when target is moving, maybe PPC hits should freeze my convergence for a bit, maybe my damage beyong optimal range will be weaker, but my infra-red sensors still are going to detect distance of a target and converge on it. In no way I'm gonna stand for enemies being transparent to them for no reason. There's hundreds of more sensible ways to encourage scouting.

[Tahawus]

rook, on 16 September 2015 - 07:15 AM, said:

Couldn't you also put the convergence distance super close, something like 100m? Weapons would still diverge at far distances.

I don't think that would be a good idea at all.

I'd far rather have it fire straight out, than have my beams cross in front of my nose at a relatively steep angle.

[pyrocomp]

DivineEvil, on 16 September 2015 - 07:32 AM, said:

I'm all for convergence mechanics - just another way to balance mechs around each-other. Nevertheless, my weapons must converge at whatever I'm pointing at no matter what it is. Maybe it might take a couple of seconds, maybe my fire will not be as accurate when target is moving, maybe PPC hits should freeze my convergence for a bit, maybe my damage beyong optimal range will be weaker, but my infra-red sensors still are going to detect distance of a target and converge on it. In no way I'm gonna stand for enemies being transparent to them for no reason. There's hundreds of more sensible ways to encourage scouting.

You never inquired how the range finder works and how it can be rendered useless? Same with whatever sensors. The fixed convergence at an optimum distance for each weapon is the simplest way to deal with too high long range alphas and addressing scouting. If there are other way of addressing scouting and lowering spam that you can stand we will surely be interested in hearing about those.